Gene Replacement and Fluorescent Labeling to Study the Functional Role of Exopolysaccharides in Bifidobacterium animalis subsp. lactis

micb-08-01405 July 21, 2017 Time: 18:28 # 1 ORIGINAL RESEARCH published: 25 July 2017 doi: 10.3389/ micb.2017.01405 Edi ed by: Rebeca Ma ín, INRA-Cen e Jouy-en-Josas, F ance Re iewed by: Analia G aciela Ab aham, Cen o de In es igacion y Desa ollo en C io ecnologia de Alimen os, A gen ina Melinda J. Maye , Ins i u e o Food Resea ch, Uni ed Kingdom *Co espondence: Pa icia Ruas-Madiedo [email p o ec ed] Special y sec ion: This a icle was submi ed o Food Mic obiology, a sec ion o he jou nal F on ie s in Mic obiology Recei ed: 07 June 2017 Accep ed: 11 July 2017 Published: 25 July 2017 Ci a ion: Cas o-B a o N, Hidalgo-Can ab ana C, Rod iguez-Ca ajal MA, Ruas-Madiedo P and Ma golles A (2017) Gene Replacemen and Fluo escen Labeling o S udy he Func ional Role o Exopolysaccha ides in Bi idobac e ium animalis subsp. lac is. F on . Mic obiol. 8:1405. doi: 10.3389/ micb.2017.01405 Gene Replacemen and Fluo escen Labeling o S udy he Func ional Role o Exopolysaccha ides in Bi idobac e ium animalis subsp. lac is Nu ia Cas o-B a o1, Claudio Hidalgo-Can ab ana1, Miguel A. Rod iguez-Ca ajal2, Pa icia Ruas-Madiedo1*and Abela do Ma golles1 1Depa men o Mic obiology and Biochemis y o Dai y P oduc s, Ins i u o de P oduc os Lác eos de As u ias – Consejo Supe io de In es igaciones Cien í icas, Villa iciosa, Spain, 2Depa men o O ganic Chemis y, Uni e sidad de Se illa, Se illa, Spain An ex acellula laye o exopolysaccha ides (EPS) co e s he su ace o some Bi idobac e ium animalis subsp. lac is s ains, which could be o ele ance o i s p obio ic pe o mance. In o de o unde s and he unc ional cha ac e is ics o B. animalis subsp. lac is, wo isogenic s ains ha di e in hei EPS-p oducing pheno ype, due o a single mu a ion in he gene Bala _1410, we e s udied. By means o a double c osso e ecombina ion s a egy, success ully used o he i s ime in bi idobac e ia, Bala _1410 in he ype s ain B. animalis subsp. lac is DSM10140 was eplaced by a mu a ed gene con aining a non-synonymous mu a ion p e iously associa ed wi h he appea ance o a mucoid- opy pheno ype. Nuclea magne ic esonance and SEC-MALS analyses showed ha he no el s ain ha bo ing he mu a ion acqui ed a opy pheno ype, due o he p oduc ion o a high molecula weigh (HMW)- EPS ha is no p oduced in he wild- ype s ain. Fluo escence labeling o bo h s ains wi h wo luo escen p o eins, m-Che y and G een Fluo escen P o ein, was achie ed by exp essing he co esponding genes unde he con ol o a na i e selec ed p omo e ( he elonga ion ac o Tu p omo e ). Rema kably, quali a i e and quan i a i e luo escence analyses demons a ed ha he opy s ain displays a lowe capabili y o adhe e o human in es inal epi helial cells. In addi ion, he p esence o he HMW-EPS educed he capabili y o he p oducing s ain o o m bio ilms upon h ee di e en abio ic su aces. This wo k also highligh s he ac ha di e en EPS con e a iable unc ional cha ac e is ics o he bi idobac e ial su ace, which may be ele an o he pe o mance o B. animalis subsp. lac is as a p obio ic. The cons uc ion o molecula ools allowing he unc ional cha ac e iza ion o su ace s uc u es in nex gene a ion p obio ics is s ill a challenging issue ha dese es u he a en ion, gi en he ele an ole ha such molecules mus play in he in e ac ion wi h he hos . Keywo ds: Bi idobac e ium, exopolysaccha ide, gene eplacemen , luo escen p o eins, NMR, SEC-MALS, bio ilms F on ie s in Mic obiology | www. on ie sin.o g 1July 2017 | Volume 8 | A icle 1405 micb-08-01405 July 21, 2017 Time: 18:28 # 2 Cas o-B a o e al. Role o Bi idobac e ial EPS on Adhesion INTRODUCTION The de ini ion o a p obio ic, p oposed in 2001 by FAO/WHO, s a es ha is “li e mic oo ganisms which when adminis e ed in adequa e amoun s con e a heal h bene i on he hos .” The mos commonly comme cialized p obio ics a e some species om Bi idobac e ium and Lac obacillus gene a ha ha e been accep ed as sa e due o hei long his o y o use and hey a e o en deli e ed in o ood o mula ions (Hill e al., 2014). Nowadays, i is becoming mo e e iden ha ce ain in es inal commensal mic oo ganisms could be bene icial o co ec mic obial dysbioses ha ha e been ela ed wi h some heal h diso de s; howe e , hey ha e no been used o p omo e heal h ye and, in case hey will be applied in his con ex , hey would be ea ed as no el d ugs mo e han ood supplemen s. These mic oo ganisms can be conside ed as “nex gene a ion p obio ics” (NGP) and hey a e e med as “li e bio he apeu ic p oduc s” (LBP) in he new egula o y amewo k o he Food and D ug Adminis a ion (FDA) o Uni ed S a es o Ame ica (O’Toole e al., 2017). Some o he p oposed NGP belong o gene a Akke mansia,Bac e oides, and Faecalibac e ium. O ally deli e ed p obio ics es ablish he main con ac poin wi h he hos a he in es inal mucosa le el; in his loca ion, he p obio ic-mic obio a-cell in e play will d i e posi i e physiological bene i s. Despi e as esea ch e o s made in o he mechanisms behind he bene icial e ec s, he manne o p obio ic ac ion s ill emains unclea (Ga eau e al., 2010;Wan e al., 2015). The ( ansi o y) con ac be ween bac e ia and in es inal epi helial cells migh be ele an o ini ia e his in e cellula dialog; he su ace mic obial associa ed molecula pa e ns (MAMPs) in e ac ing wi h he hos pa e n ecogni ion ecep o s (PRR) a e in ol ed in igge ing he cellula esponse (Lebee e al., 2010b;Wes e mann e al., 2016). One o he mos ex e nal laye s co e ing he bac e ial su ace is cons i u ed by exopolysaccha ides (EPS), which a e ca bohyd a e polyme s whose gene ic de e minan s a e p esen in in es inal bac e ia, including mos species o he genus Bi idobac e ium (Fe a io e al., 2016). In ac , some o he bene icial p ope ies a ibu able o he p oducing bi idobac e ia ha e been associa ed wi h hei EPS and hei physical-chemical cha ac e is ics (Hidalgo-Can ab ana e al., 2014, 2016;Schia i e al., 2016). Addi ionally, i has been p o en ha EPS p oduced by some NGP, such as Faecalibac e ium p ausni zii, also has an i- in lamma o y p ope ies in i o, hus his bac e ium is being p oposed as he apeu ic agen o ea in es inal in lamma o y p ocesses (Rossi e al., 2015). The de elopmen o ools allowing he s udy o he mechanisms o ac ion, ei he o well ecognized p obio ics o NGP, is essen ial in o de o choose hose s ains which a e mo e aluable o each a ge popula ion and heal h bene i . One o he app oaches is he use o ec o s con aining di e en labeling sys ems; hose applied o lac ic acid bac e ia and bi idobac e ia ha e ecen ly been e iewed (Lande e e al., 2016). A luci e ase- based epo e sys em was de eloped o moni o he pe o mance o Bi idobac e ium b e e UCC2003 unde in i o condi ions (C onin e al., 2008). Di e en luo escen p o eins ha e been success ully used as well o label bi idobac e ia. In his way, B. b e e,B. longum subsp. Longum, and B. bi idum we e labeled wi h cyan luo escen p o ein (CFP), g een luo escen p o ein (GFP), yellow luo escen p o ein (YFP) o mChe y unde he p omo e o he gap gene (Pgap) o B. bi idum (G imm e al., 2014). Addi ionally, a GFP luo escen p o ein con aining a la in-mono-nucleo ide-based co ac o (e oglow-Pp1), which emi s luo escence in he p esence/absence o oxygen, was included in a ec o unde con ol o he elonga ion ac o Tu (P u om B. longum) ha eplica es in B. longum and B. b e e (Lande e e al., 2014). This sys em, which supposes an ad an age o s udy bi idobac e ia in s ic anae obic condi ions, was la e ly alida ed unde he con ol o o he p omo e s (Mon eneg o- Rod íguez e al., 2015). Bi idobac e ium animalis subsp. lac is is one o he mos widely used p obio ics and he e a e se e al human in e en ion s udies suppo ing i s bene icial e ec s (Tojo e al., 2014). F om an indus ial poin o iew, i is one o he mos obus bi idobac e ial species which acili a es i s inclusion in oods o ood supplemen s (Bogsan e al., 2014). The aims pu sued in he cu en wo k we e: (i) o ob ain an EPS-p oducing a ian by means o a double c osso e ma ke -less s a egy, which p oduces a ch omosomally s able new a ian , (ii) he cons uc ion o EPS-p oducing B. animalis subsp. lac is s ains ha bo ing luo escen p o eins, which ha e no been epo ed in li e a u e o da e, and (iii) he demons a ion ha di e en EPS ha e an in luence on he in e ac ion o he p oducing s ain wi h bio ic and abio ic su aces. To achie e hese goals, a model o B. animalis subsp. lac is s ains, which p oduced EPS wi h di e en physical-chemical cha ac e is ics, was ini ially used. This model was p e iously de eloped o demons a e ha a single mu a ion in he gene Bala _1410, coding o a p o ein in ol ed in he elonga ion o he polyme chain, was di ec ly ela ed o a highe abundance o he high molecula weigh (HMW)-EPS ac ion (abou 106Da) ha con e ed a opy- mucoid pheno ype o he p oducing s ain (Hidalgo-Can ab ana e al., 2015). Indeed, s ains p oducing HMW-EPS a e able o a enua e he immune esponse (López e al., 2012) and hey ha e been p oposed o hei applica ion in educing in es inal in lamma o y s a es (Hidalgo-Can ab ana e al., 2016). MATERIALS AND METHODS Bi idobac e ia S ains, Plasmids and Cul u e Condi ions The B. animalis subsp. lac is and Esche ichia coli s ains, as well as he plasmids and oligonucleo ides used in his s udy, a e lis ed in Table 1.E. coli DH11S (In i ogenTM, The mo- Fishe Scien i ic Inc., Wal ham, MA, Uni ed S a es) was g own in Lu ia-Be ani (LB) b o h a 37◦C unde shaking condi ions (200 pm). Bi idobac e ial s ains we e cul i a ed in MRSc [MRS (Bioka Diagnos ics, Beau ais, F ancia) supplemen ed wi h 0.25% L-cys eine-HCl (Sigma-Chemical Co., S . Louis, MO, Uni ed S a es)] a 37◦C unde anae obic condi ions (80% N2, 10% CO2, 10% H2) in a MG500 chambe (Don Whi ley Scien i ic, Wes Yo kshi e, Uni ed Kingdom). Bac e ial cul u es and compe en cells we e p epa ed unde s anda dized condi ions F on ie s in Mic obiology | www. on ie sin.o g 2July 2017 | Volume 8 | A icle 1405 micb-08-01405 July 21, 2017 Time: 18:28 # 3 Cas o-B a o e al. Role o Bi idobac e ial EPS on Adhesion TABLE 1 | Bac e ial s ains, plasmids, and oligonucleo ide p ime s used in his s udy. S ains Desc ip ion Re e ence E. coli DH11S mc A1(m -hsdRMS-mc BC)1(lac-p oAB)1( ec1398)deoR psL s l- hi-F’ p oAB+lacIqZ1M15 In i ogen B. animalis subsp. lac is DSM10140TType s ain, Plasmid ee, EPS+, no opy pheno ype DSMZ collec ion (yogu isola ed) IPLA-R1 Plasmid ee, EPS+, opy-mucoid pheno ype IPLA collec ion (bile-sal adap ed) DSM10140-1Bala _1410 DSM10140 lacking he gene Bala _1410, no opy pheno ype Hidalgo-Can ab ana e al., 2015 DSM10140-Bala _1410S89L (=S89L) DSM10140 mu an ob ained a e gene in eg a ion o Bala _1410S89L, EPS+, opy-mucoid pheno ype This wo k DSM10140-mChe y DSM10140 ha bo ing pCAS-mChe y This wo k DSM10140-GFP DSM10140 ha bo ing pCAS- GFP This wo k S89L-mChe y DSM10140-Bala _1410S89L ha bo ing pCAS-mChe y This wo k S89L-GFP DSM10140-Bala _1410S89L ha bo ing pCAS- GFP This wo k Plasmids Desc ip ion Re e ence pJL74aE. coli-Bi idobac e ium cloning ec o ; Amp Sp ; in eg a i e, non- eplica i e in Bi idobac e ium Hidalgo-Can ab ana e al., 2015 pJL-ups /Bala _1410S89L /ds / (=pCHC3) pJL74 con aining Bala _1410S89L oge he wi h he ups eam (3 kb) and downs eam (2.7 kb) egions This wo k pAM1aE. coli-Bi idobac e ium cloning ec o ; Amp Em Al a ez-Ma ín e al., 2008 pCAS-mChe y mChe y luo escen p o ein gene used o he elonga ion ac o Tu p omo e (P u )o B. animalis subsp. lac is in pAM1 This wo k pCAS- GFP GFP luo escen p o ein gene used o he elonga ion ac o Tu p omo e (P u )o B. animalis subsp. lac is in pAM1 This wo k Oligonucleo ides Sequence (50–30) Re e ence Bala _1410 -GR-FbTATATAGGGCCCGTCACCTCGTCACCATGAGCbHidalgo-Can ab ana e al., 2015 Bala _1410 -GR-RbTATATAAGATCTCCACGAGAGCACACGAAGAC Hidalgo-Can ab ana e al., 2015 In-Bala _1410 -F GGTATGATGTGCAGATTCGGCTTC This wo k In-Bala _1410 -R TACATGGCCGAGAACGAGGTAAACC This wo k Spec-F GGAGAAGATTCAGCCACTGC Hidalgo-Can ab ana e al., 2015 Spec-R TTAGTCGTCGTATCTGAACC Hidalgo-Can ab ana e al., 2015 PTu_FbTATATAAAGCTTACATCCGTTACGAATCACGC This wo k mCh_RbTATATATCTAGATTATTACTTGTACAGCTCGTCC This wo k GFP_RbTATATATCTAGATTATTATTTGTATAGTTCATCC This wo k PTu_mCh_FcGTCCAGGAGGACAAAAACATATGGTGAGCAAGGGCGAGG This wo k mCh_PTu_RcCCTCGCCCTTGCTCACCATATGTTTTTGTCCTCCTGGAC This wo k PTu_GFP_FcGTCCAGGAGGACAAAAACATATGCGTAAAGGAGAAGAAC This wo k GFP_PTu_RcGTTCTTCTCCTTTACGCATATGTTTTTGTCCTCCTGGAC This wo k aAmp , Em , and Sp , esis ance o ampicillin, e y h omycin, and spec inomycin, espec i ely. bRes ic ion enzyme si es a e unde lined. cComplemen a y sequences o splicing o e lap ex ension PCR a e bold ma ked. (Hidalgo-Can ab ana e al., 2015) and ampicillin (100 µg/ml), spec inomycin (100 µg/ml) o e y h omycin (2.5 µg/ml) we e added when equi ed (Table 1). Molecula Techniques Isola ion o Ch omosomal and Plasmid DNA, and Plasmid Manipula ion Ch omosomal DNA om B. animalis subsp. lac is was isola ed using he GeneElu eTM Bac e ial Genomic DNA ki (Sigma–Ald ich, Do se , Uni ed Kingdom). Plasmid DNA was isola ed om E. coli using he Qiagen Plasmid Midi ki (Qiagen, Hilden, Ge many), whe eas he plasmid isola ion om ecombinan bi idobac e ia was pe o med by means o he GeneElu eTM Plasmid Minip ep ki (Sigma–Ald ich). Manu ac u e ’s ecommenda ions we e ollowed in bo h cases. Fo bi idobac e ial s ains, lysozyme (9 mg/ml, Me ck, Da ms ad , Ge many) and mu anolysin (5U, Sigma–Ald ich) we e added du ing he lysis s ep ollowed by incuba ion a 37◦C F on ie s in Mic obiology | www. on ie sin.o g 3July 2017 | Volume 8 | A icle 1405 micb-08-01405 July 21, 2017 Time: 18:28 # 4 Cas o-B a o e al. Role o Bi idobac e ial EPS on Adhesion o 1 h. DNA concen a ion was measu ed in Gene5TM Teck3 Module (BioTek, Ve mon , Uni ed S a es). Fo plasmid cons uc ions, PCRs we e pe o med using Pla inumR P x DNA Polyme ase (In i ogenTM). Diges ions and liga ions we e made wi h es ic ion endonucleases om Taka a (Taka a Bio G oup, O su, Japan) and wi h T4 DNA ligase om In i ogenTM, espec i ely. All eagen s we e used acco ding o he manu ac u e s’ ins uc ions. PCR p oduc s we e checked by elec opho esis in TAE bu e [40 mM TRIS, 20 mM ace ic acid, 1 mM EDTA (pH 8)] on 1% aga ose gels and hen s ained wi h e hidium b omide (0.5 µg/ml). DNA pu i ica ion om he aga ose gels was pe o med using QIAquick Gel Ex ac ion Ki (QIAgene) and sequenced a Mac ogen Inc. (Seoul, Sou h Ko ea). BLAST algo i hm was used o sequence simila i y analysis. Finally, Eu x-Taq DNA Polyme ase om Roboklon GmbH (Be lin, Ge many) was used o check plasmid cons uc ions in E. coli and plasmid in eg a ion in he bi idobac e ial ch omosome. Gene Replacemen : Plasmid Cons uc ion and Double C osso e E en s The ch omosomal DNA om B. animalis subsp. lac is IPLA-R1 was used as a empla e o PCR ampli ica ion using he speci ic p ime s Bala _1410-GR-F/R (Table 1) o he cons uc ion o he plasmid o gene eplacemen . These p ime s ampli y 7.1 kb which con ain he Bala _1410S89L gene (Hidalgo-Can ab ana e al., 2015) and i s lanking egions: ups eam (3 kb) and downs eam (2.7 kb). The PCR p oduc was diges ed wi h ApaI and BglII and cloned in o pJL74 p e iously diges ed wi h he same enzymes. Liga ion was pe o med o e nigh a 16◦C and he liga ion mix u e was pu i ied and ans o med in o E. coli DH11S elec ocompe en cells. The esul ing plasmid was named pCHC3 (pJL-ups /Bala _1410S89L/ds ) and was in oduced in o B. animalis subsp. lac is DSM10140- 1Bala _1410 elec ocompe en cells p epa ed as p e iously epo ed (Hidalgo-Can ab ana e al., 2015). A e ans o ma ion, bi idobac e ial cells we e immedia ely eco e ed in 2 ml o MRSc and incuba ed a 37◦C unde anae obic condi ions o 4–6 h be o e pla ing on o he same aga -medium con aining spec inomycin (100 µg/ml). Pla es we e hen incuba ed o 48 o 72 h a 37◦C in anae obic condi ions. T ans o man s we e checked o plasmid in eg a ion in o he ch omosome (single c osso e ) by PCR using he speci ic p ime s In-Bala _1410- F/R and Spec-F/R (Table 1). A his poin o he expe imen , wo o he checked colonies acqui ed he isually ecognizable opy pheno ype (ha ing he in eg a ed plasmid) and hey we e selec ed o be g own in 10 ml MRSc wi hou an ibio ic. Two subcul u es (pe day) we e made o 5 days o o ce he loss o he plasmid (second c osso e ) and, a e wa d, hese bac e ial cul u es we e pla ed on o aga -MRSc wi hou an ibio ics o 48 h. Se e al colonies we e picked up and each o hem was g own in aga -MRSc, wi h and wi hou spec inomycin, o selec he non-an ibio ic esis an colonies due o he loss o he plasmid. These colonies we e checked by PCR using he speci ic p ime s In-Bala _1410-F/R and Spec-F/R o analyze he p esence o Bala _1410S89L and he absence o spec ynomicin- esis ance genes in o he ch omosome; besides, he opy pheno ype, associa ed wi h he p esence o Bala _1410S89L, was use ul o he selec ion o he igh colonies. Thus, one s ain wi h a opy cha ac e , hen pu a i ely ca ying he Bala _1410S89L gene, and being sensi i e o spec ynomicin, was selec ed and named DSM10140-Bala _1410S89L, o S89L in i s abb e ia ed o m (Table 1). To con i m i s gene ic backg ound, he ch omosomal DNA om S89L was ob ained and an inne agmen (1 kb) o Bala _1410S89L gene, con aining he mu a ion (C o T ansi ion) esponsible o he opy ai (Hidalgo-Can ab ana e al., 2015), was ampli ied using he In-Bala _1410-F/R p ime s. The gene ic backg ound o he eps clus e su ounding he inse ion and he comple e inse was also checked wi h a se o p ime s ha ampli y 1 kb o e lapping agmen s (da a no shown). All hese PCR p oduc s we e sequenced a Mac ogen Inc. o con i m he absence o undesi able mu a ions. S ain Labeling Using Fluo escence Plasmids Plasmids ha bo ing luo escen p o eins unde he con ol o he elonga ion ac o Tu p omo e om B. animalis subsp. lac is we e cons uc ed using splicing o e lap ex ension PCR s a egy o use he DNA sequences (Vallejo e al., 1994). The “elonga ion ac o Tu” p omo e was ampli ied om he ch omosomal DNA o DSM10140 s ain using he speci ic p ime s PTu_F/mCh_PTu_R (Table 1). The gene encoding mChe y luo escen p o ein was ampli ied om he pVG-mChe y plasmid (G imm e al., 2014) wi h speci ic p ime s PTu_mCh_F/mCh_R. The PCR p oduc s ha ha e complemen a y ails, be ween each o he , we e size- checked in 1% aga ose gel. Then, splicing o e lap ex ension PCR was pe o med o use bo h agmen s. The same p o ocol was ollowed o use he elonga ion ac o Tu p omo e o GFP gene: he Tu p omo e was ampli ied using speci ic p ime s PTu_F/GFP_PTu_R, and GFP gene was ob ained om he pVG-GFP plasmid (G imm e al., 2014) using speci ic p ime s PTu_GFP_F/GFP_R. Fused DNA agmen s we e checked by elec opho esis and pu i ied om aga ose gels. The used DNA agmen s we e diges ed wi h HindIII and XbaI a 37◦C o 3 h, as well as he plasmid pAM1 (Al a ez-Ma ín e al., 2008) which was also dephospho yla ed. Diges ions we e also pu i ied om aga ose gels and used o pe o m o e nigh liga ions a 4◦C. Elec ocompe en E. coli DH11S cells we e ans o med wi h he liga ion mix u es and selec ion o clones was pe o med by adding ampicillin (100 µg/ml) o he cul u e medium. The esul ing plasmids we e named pCAS-mChe y and pCAS-GFP (Table 1). The s ains B. animalis subsp. lac is DSM10140 and S89L we e ans o med wi h hese plasmids and ou luo escen clones we e selec ed by adding e y h omycin (2.5 µg/ml) o he cul u e medium; he ecombinan s ains we e named as DSM10140-mChe y, DSM10140-GFP, S89L-mChe y and S89L- GFP (Table 1). Quali a i e and Quan i a i e Fluo escence De ec ion Fluo escence Scanning The luo escen bi idobac e ial s ains we e g own on o he su ace o aga -MRSc con aining e y h omycin, o 3 days a 37◦C unde anae obic condi ions. The luo escence o he F on ie s in Mic obiology | www. on ie sin.o g 4July 2017 | Volume 8 | A icle 1405 micb-08-01405 July 21, 2017 Time: 18:28 # 5 Cas o-B a o e al. Role o Bi idobac e ial EPS on Adhesion colonies was checked in he Typhoon 9400 scanne (GE Heal hca e, Biosciences, Uppsala, Sweden). GFP was exci ed wi h blue lase (488 nm) and emission was de ec ed wi h 526 nm band- pass il e . In he case o mChe y, exci a ion was pe o med wi h ed lase (633 nm) and emission was acqui ed wi h 580 nm band- pass il e . These pla es we e scanned a a esolu ion o 100 µm pixel size. Fluo escence Mic oscopy and Con ocal Scanning Lase Mic oscopy (CSLM) To isualize he bi idobac e ia exp essing he luo escen p o eins, o e nigh g own cul u es (in MRSc +e y h omycin) we e washed, placed on a slide co e ed wi h co e slip No.1 (0.13–0.16 mm hick). These p epa a ions we e obse ed wi h he Leica DMi8 in e ed mic oscope (Leica Mic osys ems GmbH, Heidelbe g, Ge many), using a 100×oil imme sion objec i e. The FITC il e cube (exci a ion 480/40, emission 527/30) and RHOD il e cube (exci a ion 546/10, emission 585/40) we e used o isualiza ion o he bi idobac e ia ha bo ing GFP o mChe y p o eins, espec i ely. Fluo escen (mChe y) bi idobac e ia adhe ed on he op o he in es inal cell line HT29 o in o glass slides (as will be desc ibed nex ) we e isualized wi h he Leica TCS AOBS SP8 X con ocal in e ed mic oscope [Ex e nal Se ice Uni (ESU) o he Uni e si y o O iedo, As u ias, Spain]. To isualize DAPI luo och om, samples we e exci ed a 405 nm, by a blue- iole lase diode, whe eas o de ec he mChe y hey we e exci ed a 587 nm by a whi e ligh lase . Z-s acks o HT29 monolaye s o bi idobac e ial bio ilms upon glass µ-slides we e acqui ed wi h a 63×/1.4 oil objec i e. When needed, a 2.50 op ical zoom was used o acqui e images o de ailed egions. Image cap u es we e eo de ed and p ocessed wi h he Leica Applica ion Sui X so wa e ( e sion 1.8.1.13759, Leica). Fluo escence Spec ome y Fluo escence quan i ica ion was pe o med wi h o e nigh cul u es o he ou luo escen bi idobac e ia, as well as he wo pa en al DSM10140 and S89L s ains used as nega i e con ols. Cells we e washed wi h PBS and s anda dized o an equal OD600 nm; addi ionally, hey we e pla ed in he co esponding (wi h and wi hou an ibio ic) aga -MRSc media. A e wa d, he s anda dized bac e ial suspensions we e 10- old concen a ed and om hem, se ial (hal ) dilu ions we e p epa ed in PBS. 96-well Lumi ackTM 600 whi e polys y ene pla es (VWR, Radno , PA, Uni ed S a es) we e illed wi h 200 µl (pe well) o each bi idobac e ial dilu ion. Fluo escence was measu ed on he Ca y Eclipse (Va ian Ibé ica, S.A. Mad id, Spain) luo escence spec ome e using he ollowing condi ions: 470 nm exci a ion/525 nm emission, o GFP quan i ica ion, and 585 nm exci a ion/610 nm emission o mChe y quan i ica ion. The co esponding luo escence backg ound, de e mined om he con ol samples (pa en al, non-labeled bi idobac e ial suspensions), was sub ac ed om da a ob ained o he luo escen s ains. This expe imen was pe o med wi h h ee biological eplica es. Finally, linea eg ession equa ions be ween he luo escence emi ed and he numbe o bac e ia (Log CFU/ml) we e calcula ed, as well as he co esponding coe icien o de e mina ion (R2) ha shows how well he da a i s o he linea eg ession. Flow Cy ome y Fluo escence o bi idobac e ial suspensions, ob ained as p e iously desc ibed, we e also quan i ied in he Cy omics FC500 (Beckman Coul e , Ba celona, Spain) loca ed in he ESU om he Uni e si y o O iedo. A ix acquisi ion ime o 90 s wi h “hi” acquisi ion speed was used. The 488 nm lase was applied o he exci a ion o bo h luo och omes and he selec ion o he bi idobac e ial popula ion was made by means o FSC log/SSC log (size/complexi y). This ga e was used o plo he FL1 ( o GFP de ec ion) s. FL3 ( o mChe y de ec ion) his og ams; he il e s 525/40 and 620/30 we e used o he de ec o s FL1 and FL3, espec i ely. The absence o au o- luo escence was checked in he co esponding non-labeled bac e ia (Supplemen a y Figu e S1). In he labeled s ains he eco ded luo escence was compensa ed o a oid he o e lapping o bo h luo och omes. Finally, se ial dilu ions o he samples ( om 1/2 o 1/100) we e measu ed and he linea eg ession equa ions be ween he “ luo escence emi ed” ( o al numbe o e en s mul iplied by he mean luo escence in ensi y) and he numbe o bac e ia (CFU/ml); he R2coe icien s we e calcula ed as well (Supplemen a y Figu e S2). Chemical Analysis o Pu i ied EPS EPS Pu i ica ion The EPS om s ains DSM10140, S89L and IPLA-R1 we e isola ed om he bi idobac e ial biomass collec ed wi h wa e om he su ace o aga -MRSc pla es as p e iously desc ibed (Ruas-Madiedo e al., 2010). Each bac e ial suspension was mixed wi h 1 olume o 2 M NaOH and kep o e nigh a oom empe a u e unde mild shaking. Bac e ia we e elimina ed by cen i uga ion and he EPS om he supe na an was p ecipi a ed wi h wo olumes o chilled absolu e e hanol o 48 h a 4◦C. P ecipi a ed sedimen was collec ed wi h ul a-pu e wa e and dialyzed agains wa e , using dialysis ubes o 12–14 kDa molecula mass cu o (Sigma), a 4◦C o 3 days wi h a daily change o wa e . Finally, each dialyzed sample was eeze-d ied in o de o ob ain he c ude-EPS ma e ial om each s ain. To isola e he HMW EPS- ac ion om s ains S89L and IPLA-R1, he c ude-EPS (25 mg) was dissol ed in ul a-pu e wa e (10 ml), dialysed (agains wa e , o 72 h a 4◦C) using Spec a/Po Floa - A-Lyse 100 kDa MWCO ubes (Sigma), and he con en o hese dialyzed ubes was eeze-d ied (Lei e s e al., 2011). SEC-MALLS Analysis The mola mass dis ibu ion o he c ude-EPS and HMW- EPS, as well he quan i ica ion o he ela i e amoun o he di e en size- ac ions, we e pe o med by means o size exclusion ch oma og aphy (SEC); a ch oma og aphic sys em (Wa e s, Mil o d, MA, Uni ed S a es) coupled in se ies wi h a e ac i e index (RI) de ec o (Wa e s) and wi h a mul i-angle lase ligh sca e ing de ec ion (MALLS, Dawn Heleos II, Wya Eu ope GmbH, Dembach, Ge many) was used as p e iously desc ibed (Nikolic e al., 2012). F on ie s in Mic obiology | www. on ie sin.o g 5July 2017 | Volume 8 | A icle 1405 micb-08-01405 July 21, 2017 Time: 18:28 # 6 Cas o-B a o e al. Role o Bi idobac e ial EPS on Adhesion NMR Analysis The HMW-EPS ac ions we e analyzed by nuclea magne ic esonance (NMR) a he acili ies o he Uni e si y o Se ille (Se ille, Spain). A sample o 10 mg was deu e ium-exchanged se e al imes by eeze-d ying om D2O and hen examined in solu ion (10 mg/750 mL o 99.96% D2O, Sigma–Ald ich). Spec a we e eco ded on a B uke AV500 spec ome e (B uke BioSciences, Mad id, Spain) ope a ing a 500.13 MHz (1H). Chemical shi s we e gi en in ppm, using he HDO signal (4.31 ppm a 343 K) as e e ence (Go lieb e al., 1997). The 2D he e onuclea one-bond p o on-ca bon co ela ion expe imen was egis e ed in he 1H-de ec ion mode ia single-quan um cohe ence (HSQC). A da a ma ix o 256 ×1K poin s was used o digi ize a spec al wid h o 5208 in F2 and 22522 Hz in F1. 13C decoupling was achie ed by he GARP scheme. Squa ed-cosine- bell unc ions we e applied in bo h dimensions, and ze o- illing was used o expand he da a o 1K ×1K. Adhesion o HT29 The in es inal epi helial cell line HT29 (ECACC 91072201, Eu opean Collec ion o Cell Cul u es, Salisbu y, Uni ed Kingdom) was used o es he adhesion capabili y o he luo escence-labeled and non-labeled bi idobac e ial s ains; B. animalis subsp lac is BB-12 was used as e e ence s ain. HT29 was main ained unde s anda d condi ions using McCoy’s medium (MM, Sigma) supplemen ed wi h 10% e al bo ine se um (Sigma) and wi h a mix u e o an ibio ics (50 µg/ml penicillin, 50 µg/ml s ep omycin, 50 µg/ml gen amicin and 1.25 µg/ml ampho e icin B, Sigma). The adhesion expe imen s we e pe o med upon 11-day old HT29 monolaye s g own in mic o i e pla es. Bi idobac e ial suspensions, p epa ed in MM (wi hou an ibio ics), we e added o each well a a io 10:1 (bi idobac e ia: HT29) and incuba ed o 1 h a 37◦C/5% CO2 (Nikolic e al., 2012). Fo he non-labeled s ain he numbe o bac e ia added and bac e ia adhe ed was de e mined by pla ing on aga -MRSc and he adhesion pe cen age was calcula ed as he a io be ween he bac e ia adhe ed wi h espec o he bac e ia added (Nikolic e al., 2012). Fo he luo escence-labeled bac e ia, he luo escence was measu ed by means o low cy ome y, as p e iously desc ibed, and alues o absolu e luo escence we e used o p esen he adhesion esul s. In addi ion, expe imen s o compe i ion be ween he “ luo escence-labeled, opy” s ain and he “non-labeled, non- opy” s ain (o “ luo escence-labeled, non- opy” s. “non-labeled, opy”) o adhesion o HT29 we e ca ied ou ; in his case, bo h bac e ia we e added in equal amoun s o he cell line (10:1, a io bac e ia: HT29) and he absolu e luo escence was measu ed. Bi idobac e ial Bio ilm Fo ma ion upon Abio ic Su aces The capabili y o he opy and non- opy EPS-p oducing bi idobac e ial s ains o o m bio ilms was de e mined using di e en p ocedu es and abio ic su aces. A me hod, based on impedance measu emen , ecen ly desc ibed by Gu ié ez e al. (2016) o moni o in eal ime he o ma ion o bac e ial bio ilms was used. In sho , he eal ime cell analyze (RTCA) equipmen xCelligence RTCA-DP (ACEA Bioscience Inc., San Diego, CA, Uni ed S a es) was in oduced in an incuba o , a 37◦C wi h 5% CO2, a leas 2 h be o e he expe imen s. Bi idobac e ial cul u es we e washed wice wi h PBS o p epa e s anda dized suspensions in esh MRSc (∼109c u/ml) which we e placed in he wells (100 µl/well) o speci ic E-pla es (ACEA Bioscience Inc.) coa ed wi h gold-mic oelec odes ha a e able o ansmi he impedance signal. The RTCA so wa e 2.0 (ACEA Bioscience) was used o da a collec ion and he bio ilm o ma ion was ollowed o 46 h, using h ee biological eplica es o each s ain; inally, he wells we e s ained wi h c ys al iole , as will be desc ibed nex . Bio ilms we e also o med upon polys y ene pla es (96-well mic opla es Nunc, The mo-Fishe Scien i ic Inc.), upon mic oscope co e glasses (No. 1, 18 mm diame e , Ma ien eld GmbH, Lauda-Königsho en, Ge many) p e iously s e ilized by au ocla ing (121◦C, 20 min) which we e placed in o 6-well mic opla es (The mo Fishe Scien i ic Inc.), and upon he su ace o µ-slide-2-well glass bo om (Ibidi GmbH, Ma ins ied, Ge many). A e 24-h incuba ion a 37◦C in anae obic chambe , hese bio ilms we e also s ained wi h c ys al iole . Addi ionally, he luo escence-labeled bi idobac e ial bio ilms (incuba ed in da kness) o med upon co e glasses we e isualized unde he epi luo escence mic oscope and hose o med upon he su ace o µ-slide-2-well glass bo om we e de ec ed wi h he CSLM. C ys al Viole S aining The end-poin c ys al iole me hod was used o quan i y he bi idobac e ial bio ilm o ma ion upon he h ee abio ic su aces used (Gu ié ez e al., 2016). In b ie , supe na an s om di e en abio ic-ma e ial wells we e emo ed and bio ilms washed wice wi h PBS, d ied o 15 min a oom empe a u e and s ained wi h a solu ion (0.1% w/ ) o c ys al iole o 15 min. Then, bio ilms we e gen ly washed wi h wa e , de-s ained wi h a solu ion (33%) o ace ic acid o a leas 15 min and, inally, he abso bance o he supe na an s was measu ed a 595 nm in a Mic opla e Benchma k Plus (Bio-Rad, He cules, CA, Uni ed S a es) spec opho ome e . S a is ical Analysis The s a is ical package IBM SPSS S a is ics o Windows Ve sion 22.0 (IBM Co p., A monk, NY, Uni ed S a es) was used o assess di e ences among s ains by means o one-way ANOVA ollowed, when needed, by SNK (S uden -Newman– Keuls, p<0.05) mean compa ison es . The legend o each igu e indica es he analysis pe o med. Finally, he R2coe icien s, ha e lec he adjus men o linea eg ession equa ions be ween di e en pa ame e s, we e calcula ed. RESULTS AND DISCUSSION Gene a ion o a Ropy S ain wi h a Non-synonymous Mu a ion in he Gene Bala _1410 In a p e ious wo k, an isogenic mu an de i ed om he ype s ain B. animalis subsp. lac is DSM10140 by emo ing he gene Bala _1410, using a knockou mu a ion sys em based F on ie s in Mic obiology | www. on ie sin.o g 6July 2017 | Volume 8 | A icle 1405 micb-08-01405 July 21, 2017 Time: 18:28 # 7 Cas o-B a o e al. Role o Bi idobac e ial EPS on Adhesion on he in eg a i e plasmid pJL74, was cons uc ed (Hidalgo- Can ab ana e al., 2015). Ou i s aim in he p esen s udy was o ein oduce in o he genome o B. animalis subsp. lac is DSM10140-1Bala _1410 a mu a ed Bala _1410 gene con aining a non-synonymous, single nucleo ide mu a ion p e iously associa ed wi h he appea ance o a mucoid- opy pheno ype in he s ain DSM10140-1Bala _1410-pAM1-Bala _1410S89L (Hidalgo-Can ab ana e al., 2015). To do ha , a double-c osso e ma ke -less s a egy p e iously used o he dele ion o he gene, was ollowed. The s ain DSM10140-1Bala _1410 was ans o med wi h he plasmid pCHC3 (Table 1) con aining a agmen o app oxima ely 7 kb ampli ied om he genome o he s ain IPLA-R1 (Table 1), ha includes he egions immedia ely loca ed ups eam and downs eam o he gene Bala _1410 as well as he mu a ed gene Bala _1410S89L be ween hose egions. Gene in eg a ion was achie ed as p e iously desc ibed (Hidalgo- Can ab ana e al., 2015), esul ing in B. animalis subsp. lac is DSM10140-Bala _1410S89L (abb e ia ed as S89L), a s ain ha has exac ly he same gene ic backg ound as B. animalis subsp. lac is DSM10140 which unde wen a gene eplacemen ha esul ed in a C o T ansi ion in he gene Bala _1410 a posi ion 266, causing a codon change in posi ion 89 (a se ine is subs i u ed by a leucine). Al hough he e a e se e al wo ks ha ha e epo ed gene dele ion and in e up ion sys ems in bi idobac e ia, including B. b e e (Ruiz e al., 2012), B. longum (Fukuda e al., 2011;Hi ayama e al., 2012) and B. animalis subsp. lac is (A igoni and Delley, 2008;Hidalgo-Can ab ana e al., 2015), o ou knowledge his is he i s epo o a success ul gene eplacemen s a egy in bi idobac e ia. Due o he lack o gene ic ools o in oduce speci ic poin mu a ions in bi idobac e ial genomes, ou me hodology ep esen s a sui able al e na i e o o e come his limi a ion. Analysis o EPS Syn hesized by he Recombinan Ropy S ain In he EPS syn hesized by he wo opy s ains B. animalis subsp. lac is IPLA-R1 (pa en al) and S89L ( ecombinan ) he HMW-EPS ac ion (abou 1 ×106Da) was p esen in a highe p opo ion han in he non- opy DSM10140 (pa en al) s ain (Figu e 1A). I should be no iced ha he polyme ma e ial pu i ied om he h ee s ains, wi h he p ocedu es used in his s udy, is he cell-associa ed EPS bu no ha libe a ed in o he medium. P e iously, he p oduc ion o he HMW-EPS was co ela ed wi h he occu ence o he mucoid- opy appea ance in a ecombinan s ain ha bo ing he mu a ed gene in a mul i- copy plasmid (Hidalgo-Can ab ana e al., 2015); hus, cu en ly his inding was ein o ced wi h he acquisi ion o he opy pheno ype in he no el S89L ha ing he single mu a ion s abilized in o he ch omosome. A e pu i ica ion o he HMW-EPS ac ion om polyme s syn hesized by IPLA-R1 and S89L s ains, one- and wo-dimensional NMR analyses e ealed an iden ical FIGURE 1 | Physical-chemical analysis o EPS isola ed om h ee B. animalis subsp. lac is s ains. SEC-MALLS analysis o he EPS-DSM10140, EPS-S89L and EPS-IPLA-R1 showing he molecula weigh (Mw) and ela i e abundance o he high molecula weigh (HMW) ac ion in each polyme (A).1H-NMR (500 MHz, 343 K) spec a (B) and 500-MHz1H-13C- HSQC spec a (C) o HMW-EPS pu i ied om EPS-IPLA-R1 and EPS-S89L polyme s. F on ie s in Mic obiology | www. on ie sin.o g 7July 2017 | Volume 8 | A icle 1405 micb-08-01405 July 21, 2017 Time: 18:28 # 8 Cas o-B a o e al. Role o Bi idobac e ial EPS on Adhesion FIGURE 2 | Physical map o pCAS-GFP and pCAS-mChe y, in which he GFP and mChe y genes a e loca ed downs eam om he elonga ion ac o Tu p omo e (P u ) om B. animalis subsp. lac is DSM10140 (A). De ec ion o luo escence g een ( op) o ed (bo om) colonies o B. animalis subsp. lac is DSM10140-GFP and DSM10140-mChe y, espec i ely, using he Typhoon 9400 scanne luo escence scanne (le hand pho og aphs) and aspec o he colonies isualized wi h a con en ional came a ( igh hand pho og aphs) (B). Fluo escen B. animalis subsp. lac is S89L, ans o med wi h pCAS-GFP ( op) and pCAS-mChe y (bo om), isualized wi h he Leica DMi8 in e ed mic oscope using a 100×oil imme sion objec i e (C). S ain B. animalis subsp. lac is S89L-mChe y showing a colo ed pink colony wi h a opy pheno ype deno ed o he o ma ion o a long, unb eakable ilamen (D). chemical composi ion (Figu es 1B,C). These physical-chemical analyses undoub edly p o e ha he gene eplacemen s a egy applied o ob ain he ecombinan S89L s ain was success ul o in oduce he mu a ion linked o he p oduc ion o he HMW-EPS. The s uc u al epea ing uni o he HMW-EPS was al eady desc ibed o s ain IPLA-R1 (Lei e s e al., 2011); i is an hexapolysaccha ide wi h 50% hamnose con en , and i is e y simila o ha epo ed o s ain B. animalis subsp. lac is LKM512 (Uemu a and Ma sumo o, 2014). Exp ession o Fluo escen P o eins in B. animalis subsp. lac is and Fluo escence De ec ion Bi idobac e ium animalis subsp. lac is DSM10140 and S89L we e ans o med wi h he plasmids pVG-GFP and pVG-mChe y, con aining he genes coding o he p o eins GFP and mChe y, espec i ely. I was epo ed ha hese plasmids we e success ully used o he luo escen de ec ion o B. longum,B. b e e, and B. bi idum s ains, g own in simila condi ions o hose used in ou s udy (G imm e al., 2014). Howe e , no luo escence was de ec ed in ou B. animalis subsp. lac is s ains using luo escence spec ome y echniques, sugges ing ha ei he he genes a e no exp essed o he p o eins do no emi luo escence unde ou expe imen al condi ions. Since gene exp ession in he plasmids pVG-GFP and pVG-mChe y is unde he con ol o he p omo e o he glyce aldehyde-3-phospha e dehyd ogenase gene o B. bi idum (Pgap), a speci ic p omo e o B. animalis subsp. lac is was in es iga ed which could be sui able o igge he exp ession o he GFP and mChe y genes in ou s ains. Ou p e ious wo k on bi idobac e ia allowed us o depic a de ailed p o ein map o he mos abundan cy oplasmic p o eins in hei soluble p o eome (Sánchez e al., 2005, 2007). Indeed, one o he mos abundan p o eins in he soluble p o eome o some bi idobac e ia is he elonga ion ac o u ( he p oduc o he u gene; Sánchez e al., 2005;Wei e al., 2016). Fu he mo e, p e ious epo s ha e shown ha he P u o B. longum is a sui able s ong and cons i u i e p omo e able o igge he exp ession o luo escen p o eins in B. longum and B. b e e (Lande e e al., 2014). These p e ious indings indica e ha he u p omo e could be a good candida e o he exp ession o he e ologous genes in B. animalis subsp. lac is. Wi h his in mind, he p omo e o he glyce aldehyde- 3-phospha e dehyd ogenase, o iginally p esen in he plasmids F on ie s in Mic obiology | www. on ie sin.o g 8July 2017 | Volume 8 | A icle 1405 micb-08-01405 July 21, 2017 Time: 18:28 # 9 Cas o-B a o e al. Role o Bi idobac e ial EPS on Adhesion FIGURE 3 | Fluo escence quan i ica ion o GFP ( he wo le panels) and mChe y p o eins ( he wo igh panels) o he bi idobac e ial s ains ha bo ing he co esponding plasmids using he Cy omics FC500 low cy ome e ; see Supplemen a y Figu e S2 o co ela ion wi h bac e ial enume a ion in aga -MRSc (A). Linea eg ession be ween he luo escence emi ed, eco ded by he Ca yEclipse luo escence spec ome e , and he bac e ial coun s (Log CFU/ml) o se ia ed dilu ions o he luo escen bi idobac e ial suspensions; he coe icien s o de e mina ion (R2), showing how well da a i o he eg ession line equa ions, a e indica ed in bold le e s (B). pVG-GFP and pVG-mChe y, was eplaced by he ups eam egion o he u gene o B. animalis subsp. lac is DSM10140, con aining P u , yielding he plasmids pCAS-mChe y and pCAS-GFP (Table 1 and Figu e 2A). These plasmids we e used o ans o m B. animalis subsp. lac is DSM10140 and B. animalis subsp. lac is S89L; g een and ed luo escence in he esul ing s ains (DSM10140-mChe y, DSM10140-GFP, S89L- mChe y and S89L-GFP) we e de ec ed by luo escence scanning (Figu e 2B) and luo escence mic oscopy (Figu e 2C). The opy pheno ype, deno ed by he o ma ion o a ilamen , was de ec ed in he s ain S89L-mChe ey which also acqui ed a pink colo in he colony (Figu e 2D). Fu he mo e, a quan i a i e analysis o he luo escen ly labeled bi idobac e ia was pe o med using low cy ome y (Figu e 3A), and he luo escence signal was co ela ed wi h bac e ial coun s by a linea eg ession analysis (Supplemen a y Figu e S2). Ou esul s showed ha he luo escence quan i ica ion o he s ains labeled wi h he wo luo escen p o eins co ela e wi h he bac e ial coun s in aga pla es. The coe icien s o de e mina ion (R2) ob ained in hese analysis we e in all cases highe han 0.95, indica ing he sui abili y o ou app oach o quan i y luo escen ly labeled bi idobac e ial popula ions in a ange o 2 log coun uni s (Figu e 3A and Supplemen a y Figu e S2). Ve y good i s o he linea eg ession we e also ob ained in a ange o 3 log coun uni s when he luo escence quan i ica ion was ca ied ou using spec ome ic echniques (Figu e 3B). P e ious wo ks ha e shown ha o he species, such as B. longum,B. bi idum, and B. b e e, can be luo escen ly labeled wi h a a ie y o p o eins, including GFP, m-Che y, Yellow Fluo escen p o eins and Cyan Fluo escen p o eins (G imm e al., 2014;Lande e e al., 2014). Howe e , i is wo h highligh ing ha , o he bes o ou knowledge, his is he i s epo desc ibing he luo escen labeling o B. animalis subsp. lac is, which opens new possibili ies o ack he unc ional p ope ies o his Bi idobac e ium unde in i o and in i o condi ions. Rega ding he s abili y o he luo escence emission o GFP and mChe y, he luo escence signal o GFP was qui e uns able compa ed wi h ha o m-Che y. Once he cell g ow h was s opped, he cells we e no able o emi luo escence o longe han 1 h, independen ly o he echnique used o measu e he GFP luo escence. A quick dec ease in he luo escence in he wo B. animalis subsp. lac is s ains analyzed was obse ed. Howe e , m-Che y luo escence was mo e in ense and s able o longe pe iods o ime (a signi ican dec ease o he luo escen signal a e 24 h a RT was no obse ed) (da a no shown). The e o e, he s ains labeled wi h m-Che y we e selec ed o pe o m he expe imen s desc ibed in he ollowing sec ions o his wo k. F on ie s in Mic obiology | www. on ie sin.o g 9July 2017 | Volume 8 | A icle 1405